Abstract The ability of people with severe physical impairment to participate in family life, communication, work, or recreation is severely restricted without access to assistive technology (AT). Yet, as disability severity increases, so does the challenge to finding (1) an access movement that a person can perform to control AT and (2) an access technology that can detect the access movement. We propose to create a wireless movement sensing system that can learn a user?s access movement and then recognize that movement in order to wirelessly control assistive devices. Once we complete the technology development, we will measure the sensitivity and specificity of our wireless movement sensing system using the movements of ten people with SPI. We will present our results to AT experts during a focus group session and determine the perceived strengths and weaknesses of the technology. This Phase 1 research is proposed by a multidisciplinary research team consisting of AT engineers, AT clinicians, and a machine learning expert.